From control of the vehicle to personal
self-control; broadening the perspectives to driver education
M. Hatakka a,*, E. Keskinen a
, N.P. Gregersen b
, A. Glad c
, K. Hernetkoski a
a
Department of Psychology, University of Turku, 20014 Turku, Finland
b
The Swedish National Road and Transport Research Institute, 58195 Link€ooping, Sweden
c
The Institute of Transport Economics, Box 6110 Etterstad, 0602 Oslo, Norway
Received 2 July 1999; received in revised form 14 November 2000; accepted 5 July 2001
Abstract
The objective is to formulate guidelines and goals for future development in the area of driver training
and education. The content of this paper is not empirical, but merely an analytical summary or review. A
four-level descriptive model is presented in which driver behaviour is conceptualised as a hierarchy, in
which the goals and motives of the driver play an essential role. The recent constructivist ideas in mainstream
pedagogy and psychology of learning are combined with a hierarchical approach to driver behaviour.
A comprehensive framework for goals and contents of driver education (GDE framework) is
presented. Two main conclusions can be drawn. Firstly, the conceptual analysis points towards a need to
emphasise the motivational aspects in driver education more than it is done at present. Secondly, in order to
reach the goals, pedagogical methods should be re-evaluated. For example, active learning methods and use
of self-reflection should be promoted in driver education.
Ó 2002 Published by Elsevier Science Ltd.
Keywords: Driver training; Driver education; Educational methods; Theory in traffic psychology; Driver behaviour
1. Introduction
The objective is to formulate a conceptual model of driver training and education. The idea is
to be prospective, with clear emphasis on finding new goals and methods. A solid theoretical
framework is needed to cover a driver’s task as broadly as possible. Issues that are addressed
*
Corresponding author. Tel.: +358-2-333-5060; fax: +358-2-333-5063.
E-mail address: mika.hatakka@utu.fi (M. Hatakka).
1369-8478/02/$ - see front matter Ó 2002 Published by Elsevier Science Ltd.
PII: S1369-8478(02)00018-9
Transportation Research Part F 5 (2002) 201–215
www.elsevier.com/locate/trf
include, for example, what the driver has to master; which factors affect the learning process and,
in particular, which educational methods are needed.
2. Theory of traffic behaviour and driver education
2.1. Hierarchical approach to drivers’ task and skills: defining the complexity of the problem
An analysis of the driver’s task and accidents has shown that adequate psychomotor skills and
physiological functions are not sufficient for good and safe performance as a driver. This conclusion
is in line with the notion that driving is a self-paced task (N€aa€aat€aanen & Summala, 1974).
Recently, for example, Rothengatter (1997) has pointed out, that research in traffic psychology
shows not only the importance of performance factors, but also the importance of motivational
and attitudinal factors. This observation concurs with the distinction between the concepts
‘‘errors’’ and ‘‘violations’’ in driver behaviour (Parker, Reason, Manstead, & Stradling, 1995;
Reason, Manstead, Stradling, Baxter, & Campbell, 1990). Since Miller, Galanter, and Pribram
(1960), hierarchical approaches have been typical in more recent psychological attempts to explain
human behaviour. The importance of hierarchical approaches is realised also in the more general
debate in traffic psychology (Janssen, 1979; Michon, 1985, 1989; Ranney, 1994; Summala, 1985).
Although in the past hierarchical approaches have been used mainly for describing the performance
aspects of driving behaviour (Mikkonen & Keskinen, 1980; Rasmussen, 1980; Van der
Molen & B€ootticher, 1988) a hierarchical approach can also be used to combine the motivational
and attitudinal aspects of driving behaviour with performance, or operations in certain traffic
situations. The interplay between the elements of driving skill as well as the conceptualisation of
driving as a task that includes all components from motor coordination to higher motivational
levels becomes evident, when driver behaviour is schematised in a hierarchical approach
(Keskinen, 1996) (see Fig. 1).
Skills for vehicle manoeuvring and mastery of traffic situations are the basis for successful
operation in traffic and these aspects should be learned well during driver training. Psychomotor
and physiological aspects are important as basic requirements for operations at the lowest levels
of the hierarchy of driver behaviour. However, these skills are applied under guidance of higher
Fig. 1. Illustration of hierarchical levels of driver behaviour (adapted from Keskinen, 1996).
202 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
level goals and motives. This means, that in addition to the training of basic skills, driver training
should also be able to deal with these higher levels in the hierarchy and take into consideration the
driver’s goals related to driving and, for example, skills for dealing with social pressures during a
trip. Driver’s goals may have an effect on both, increase or decrease of risks. The general idea of
control of higher levels on lower levels is found also in Janssen’s (1979) and Michon’s (1985)
thinking, but they were not focussing on the question how different motivational aspects were
increasing or decreasing risks.
The idea in a hierarchical approach is, that failure as well as success at higher levels affect the
demands on skills at lower levels. A person’s general goals for life and the means for satisfying
these (e.g. developing one’s identity with car, and driving-related activities) (Gregersen & Berg,
1994; Jessor, 1987), as well as a person’s general skills for life (e.g. self-control) (Kokkonen &
Pulkkinen, 1999; Pulkkinen, 1998) can be considered as the highest level in the hierarchy.
When viewed from this perspective, it is easy to understand why several attempts to improve
safety by improving skills at the lower levels of the hierarchy only (vehicle handling skills on
slippery road) have actually failed to decrease accidents (Christensen & Glad, 1996; Glad, 1988),
exclusive vehicle handling skills of race drivers are connected with high number of accidents
(Williams & O’Neill, 1974) or that some groups benefit from training to master slippery road
conditions and other groups obtain a negative effect from it (Katila, Keskinen, & Hatakka, 1996;
Keskinen, Hatakka, & Laapotti, 1992).
The control in the hierarchy is not a simple top-down process. Changes in lower levels also have
effects on the whole system. If increased skills, or even worse, imagined increase in skills (Gregersen,
1996a) are used to satisfy needs for maintaining as high a speed as possible, the results are
very likely to be negative. If the motivational level fails to produce a safe strategy for driving, no
level of skills in mastering traffic situations or vehicle handling is high enough to compensate for
this lack of safety orientation and to result in greater safety. This is also the message of the risk
motivation theories of driver behaviour (Fuller, 1984; N€aa€aat€aanen & Summala, 1974).
Thus the view presented here considers all levels in the hierarchy to be essential for successful
operation in traffic. Even though cognitive theories of psychology allow the theoretical possibility
to use the same constructs from the bottom to the top of the hierarchy, it is obvious, that several
theories are needed to specifically understand the problems encountered at each level of the hierarchy.
We do not expect, that there would be a general traffic psychological theory that covers
all these levels, but there are for each level several theories (theories specific to traffic psychology
or theories of general psychology). Examples of those will be presented in order to illuminate the
specific problems connected with each level in the hierarchy, and to explain how each level is
present in traffic behaviour. The function of the hierarchical approach here is to make the goals of
driver training more clear. What are the aspects that driver education should include and how
these aspects can be influenced. This theoretical description of driving task should be considered
as a starting point for evaluation of specific driver education methods.
2.2. Description of hierarchical levels
2.2.1. Goals for life and skills for living
The highest level in the hierarchy refers to the motives and goals of the person in a rather broad
sense (Keskinen, 1996). Personal skills for handling different situations in life in general are also
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 203
included at this level. Traffic psychological research has indicated that general motives as well as
the developmental stage of the person are influential factors in determining traffic behaviour.
Youngsters, for example, are frequently involved in risky behaviour (Evans, 1991), but this is not
for risk in itself, but because such behaviour serves youngsters’ developmental needs. Risky behaviour
can thus be considered functional (Jessor, 1982, 1987). The effects of the motivational
aspects were emphasised by N€aa€aat€aanen and Summala (1974) when they presented their concept of
‘‘extra motives’’.
Gregersen (1996b) has emphasised the importance of lifestyle factors and values that affect
driver behaviour. Lifestyle studies (Gregersen & Berg, 1994; Schulze, 1990) show that lifestyle is
related to driving behaviour. A highly car-oriented lifestyle seems to be particularly problematic.
It can be hypothesised, that when motives for self-enhancement are realised in the traffic context,
this may increase risk. This type of information should be included in driver training in order to
make youngsters more aware of their personal processes and the possible motivational pressures
they encounter while driving.
The differences between males and females in traffic risk are aspects that can be traced back to
motivational factors. These differences are both qualitative and quantitative (Bruhning & Kuhnen,
1993; Farrow, 1987; Keskinen et al., 1992; Laapotti & Keskinen, 1998; Twisk, 1994b). Most
investigations of young drivers have, however, either focused on males or have failed to distinguish
between the behaviour of males and females (Jonah, 1990; Laapotti & Keskinen, 1998;
Renge, 1983). This is why our image of a typical young driver’s accident usually resembles the
features of male rather than female accidents. A typical image of a young driver’s accident would
be a severe single (loss of control) accident that happens when the driver is driving at high speed
during a weekend night while on a leisure trip with drinking friends.
Young female drivers’ accidents have quite different characteristics. The mechanisms behind
their accidents also seem to be different. They are so different that it may not be possible at all to
speak about typical young drivers’ accidents. For example speeding and driving while impaired
are typical for male drivers but rare for female drivers. Such differences are the result of differences
in motives (Laapotti & Keskinen, 1998). The causes of accidents seem to vary considerably, and
since the training has to be adapted to the causes, this calls for a variety of methods in training.
Effects of gender on accident involvement are probably related to the highest levels in the hierarchy,
as no suggestion has been made that females possess better technical skills for dealing
with various traffic situations or for vehicle manoeuvring than male drivers. Studies also show
that older novice drivers seem to have fewer accidents compared to young novice drivers when
mileage is controlled for (Keskinen et al., 1992, 1994). Thus, the differences have to be associated
with maturation. In addition to the effect of age, experience also has an independent effect on
accident involvement (Maycock, Lockwood, & Lester, 1991).
When a hierarchical approach is applied to driver training it points towards an emphasis on
methods that are capable of dealing with the highest levels of the hierarchy, i.e. motivational and
other factors connected more generally with driver’s strategy, motives and life skills. Modification
of behaviour is not possible without modification, or at least, awareness of personal goals.
When these aspects are considered as important targets for driver education, a need for a
change in driver training methods arises. Measures for accident prevention at the highest level of
the hierarchy focus partly on arranging support for young persons in their developmental tasks so
that they could mature in a more safe way. Even though it cannot be expected that driver edu-
204 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
cation can change a person’s life goals, education should make a person conscious of his or her
personal tendencies, that affect driving behaviour. A recent example of safety-positive trend,
which is obviously connected with the highest level of values and life goals is the radical decrease
in the licensing of youngsters in Sweden. Young people choose public transportation and are
driving less, and this has decreased accidents (Swedish Road Administration, 1998). Such trends
as choosing not to drive when other methods of transportation are available should be supported
by driver education. Perhaps, driver training should broadened into transport education.
Anticipation of risks connected with goals for life and skills for living is related to knowledge
about the relations between risk-increasing tendencies, risky motives or lifestyle and driving. The
drivers should also be aware of their own personal characteristics at this level.
There is evidence that it is possible to have at least some effects on alcohol-related violations
and other traffic violations with appropriate educational methods. In Germany, driver improvement
methods have been widely used and evaluation studies have shown reduction of recidivism
after course participation (Bartl & Stummvoll, 2000; Utzelmann & Jacobshagen, 1997).
The educational methods used in driver improvement courses are largely based on self-reflective
and self-evaluative processes and even therapeutic elements are included (Hatakka & Keskinen,
1999). The methods used in driver improvement programs are, however, not widely in use in basic
training of novice drivers (GADGET, 1999).
2.2.2. Goals and context of driving
The second highest level is related to goals and context of driving. At this level drivers decide
for what purpose, where, with whom, with what and at what time to drive. These decisions have
important consequences for traffic safety and are functions of many factors in a person’s life
situation, e.g. economical situation and in his/her personality.
This level refers partly to the navigational and planning tasks of the driver that are described in
earlier hierarchical conceptualisations (Janssen, 1979; Michon, 1985; Mikkonen & Keskinen,
1980; Van der Molen & B€ootticher, 1988), However, the view is somewhat broadened to include
also trip-related goals and driving contexts. The idea is that trip goals and driving contexts are
affected by higher level goals. While middle-aged drivers are typically involved in driving connected
with work or transportation of family members, youngsters are more often engaged in
driving for leisure purposes and in the company of friends (Laapotti, Keskinen, Hatakka, &
Katila, 1996).
Good planning of the trip may make the driving easy, while bad planning or lack of planning
can cause troubles. Proper estimation of travel time, selection of the easiest route or the most
suitable time for the trip, will help the driver to encounter less demanding traffic situations and
lessen the burden on vehicle manoeuvring skills when actually making the trip. A central aspect of
planning is, of course, the major decision to drive. These kinds of evaluations should be included
in driver training because of their direct effect on both quality and quantity of exposure and thus,
on personal risks.
Other factors that are relevant at this level are related to the social context of the driving. Social
pressure has a strong impact on driver behaviour. General social–psychological theories and
social psychological theories about traffic (Jessor, 1982, 1993; Jonah, 1986, 1990) should also be
used in the design of driver education. Passenger-related risks are emphasised in some studies
(e.g. Marthiens & Schulze, 1989). Social context in the form of peer group represents the most
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 205
important influence on the behaviour of young male drivers (Lewis, 1985). Farrow’s (1987) and
Laapotti’s (1994) results support this statement. In accidents where the cause was loss of control,
young male drivers had more passengers than females and their passengers were more likely to be
friends. Females had more often family members in their car. As Twisk (1994a) summarises: ‘‘The
young driver is not an isolated individual, but is part of a closely knit social structure. His or her
position in this social structure influences his or her current behaviour, attitudes and beliefs’’.
Again, we must point out that young drivers may be realising their personal motives through
their driving. In Vogel and Rothengatter’s (1984) study, the ‘‘pleasure in driving’’ had the highest
relative contribution to driver’s attitude towards speeding. Motivational factors, e.g. driving for
pleasure can also be seen in the selection of cars. Those male drivers who are interested in driving
fast tend to choose fast cars (Hatakka, Keskinen, Katila, & Laapotti, 1994). Those drivers who
had the highest level of ‘‘extra motives for car choice’’ (important factors in car choice were high
acceleration, powerful engine, good looking, sporty) also had the highest annual mileage and
highest number of violations even when mileage was controlled from.
Decisions made at this level of the hierarchy can be considered partly as skills in a rather
traditional sense, e.g. planning of the route and it is possible to find training methods for improving
these skills. However, such factors as risks connected with the goals of the trip and with
social pressure while driving are somewhat different aspects, and require educational methods that
help to increase subjective awareness.
Anticipation of risks at this level of the hierarchy is related to knowledge about the necessity of
planning and the problems that arise, when the driver is not able to plan driving properly.
Furthermore, the driver should be aware of the risks connected with specific goals of the trip or
driving context. If the driver is aware of typical personal goals for driving or habits when encountering
problematic situations, he or she could prepare himself or herself for dealing with the
problems.
2.2.3. Mastery of traffic situations
Mastery of traffic situations has traditionally been the central component of driver education.
The driver has to adapt his or hers individual behaviour to other road-users’ behaviour and to the
traffic environment. The driver also has to perceive and predict other road-users’ behaviour and
make his or her own behaviour predictable to others. Knowing traffic rules and behaving according
to them is one important part of these skills. In a hierarchical view, it is emphasized that
behaviour in specific traffic situations is related to the driver’s general tendencies and goals of the
trip.
Earlier hierarchical approaches (Janssen, 1979; Michon, 1985; Mikkonen & Keskinen, 1980;
Van der Molen & B€ootticher, 1988) have considered this level as a tactical level, or level of specific
traffic situations. This level is related to the negotiation of traffic situation and road designs. An
essential problem at this level for novice drivers is that insufficient skills and insufficient automatism
result in information overload and mistakes or less appropriate strategies, for example in
observation or allocation of visual attention (Mourant & Rockwell, 1972; Wikman, Nieminen, &
Summala, 1998), and longer reaction times (Quimby & Watts, 1981). Novice drivers are involved
in accidents more often as a guilty party than experienced drivers (e.g. Keskinen, 1982). There is
also a strong effect of experience on the accident liability (Maycock et al., 1991). Fuller (1984,
1988) conceptualises mastery of traffic situations as increased ability for threat avoidance and he
206 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
suggests that experience has a strong effect on this ability. Good skills for mastery of traffic situations
include, in addition to skills needed in normal situations, skills for risk-recognition in
problematic situations. Gregersen et al. (2000) results showed that increase in amount of practising
before licensing reduced novice drivers’ accident involvement. This refers to the importance
of mastery of traffic situations.
Excellent skills for mastery of traffic situations are necessarily not enough for safe driving. The
high interest in cars and driving traditionally exhibited by males does not lead to lower crash rates
even though it may lead to higher levels of skill and knowledge (Evans, 1991, p. 136). Training
courses focusing on technical mastery of traffic situations, and on producing relaxed and confident
drivers, may desensitise the driver to fear in more risky situations (Job, 1990). The increased
technical skills are likely to lead to an increase in the level of task difficulty, such as caused by
driving faster, overtaking in heavier traffic, or accepting additional secondary tasks, e.g. listening
to the radio, rather than simply to an increase in safety (Evans, 1991).
In Japan, an interesting relationship was found between the success in a driving test, the
number of hours in driving school before the driving test and accidents and violations. The less
hours of training male students had (i.e., the easier they had achieved the required level of skill)
the more often they were involved in accidents and violations. The better the males were in the
driving test the more they were involved later in accidents and violations. Both these effects remained
even when the age of males was controlled for. However, no such effect was found in
female drivers’ groups (Renge, 1983). Passing the driving test also is easier for young males in
Finland, but after licensing males have more accidents than older drivers or female drivers
(Keskinen et al., 1992). The results suggest that measurement of mastery of traffic situations that
emphasises maximal performance––which is the major topic in driving tests––is not sufficient.
Driving tests should also include elements of driving style factors. The distinction of drivers’
mistakes in traffic situations into violations, where the problem is not in basic skills but in the
intention of the driver, and slips and lapses (Reason et al., 1990) must be taken into account in
driver education design.
As a general conclusion, it can be stated that skills for mastery of traffic situations are essential
for success in traffic. It is necessary, that licensing systems should guarantee a good level of
mastery of traffic situations. However, a general problem in driver education and testing is that
training and evaluation of skills for mastering traffic situations is not enough. Driving is essentially
a self-paced task, the driver chooses the demands that are put on his or her skills. Improvement
of skills should be gained in such a way, that misuse and overestimation of personal
skills is avoided. This requires that the goals of training and the design of training methods
take into consideration both drivers’ aptitude and also drivers’ typical motivational and driving
style factors. Furthermore, awareness of personal skills and their limits should be included in
training.
2.2.4. Vehicle manoeuvring
Vehicle manoeuvring is a challenge for driver education in much the same way as skills for
coping with traffic situations. To operate a vehicle efficiently the basic manoeuvres have to be
automatized. If not, the manoeuvring will demand conscious attention and strain the limited
capacity for information processing. This leaves little capacity to observe and predict the behaviour
of other road users. This idea is well described in earlier hierarchical approaches to driver
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 207
skills (Michon, 1985; Mikkonen & Keskinen, 1980; Van der Molen & B€ootticher, 1988) and in the
literature about learning of skills (e.g. Fitts & Posner, 1967). The problems at this level are partly
related to problems of information overflow in novice drivers. During training a sufficient number
of repetition of basic manoeuvres is needed in order to achieve automaticity. In addition to the
skills needed in normal driving situations, the driver should be aware of the typical mistakes with
speed control and steering, that lead to loss of control of the vehicle. Without this knowledge
avoidance of these mistakes is not possible.
It should be noted that as for the training of skills needed in traffic situations, there are
problems connected with training of vehicle manoeuvring skills. As described above, vehiclemanoeuvring
skills are used according to the driver’s motives. Improvement of basic vehicle
handling skills will probably have positive effects on the driver’s possibilities for safe driving.
However, training of skills that are intended for use in hazardous situations seems to bear a
possibility for misuse or development of overconfidence in technical driving skills. It is important
that driver training does not create the impression that the driver’s task is mainly a manoeuvring
task. Such an impression would be an incorrect basis for the development of the driver’s behaviour.
Studies indicate that such impressions may easily be mediated by the driver training
(Gregersen, 1996a; Katila et al., 1996).
3. Combination of driver’s task and goals for learning, GDE framework
Table 1 presents a framework relating the hierarchical approach to driving with contents and
goals of driver education. Later on the framework will be referred to as ‘‘Goals for Driver Education’’
framework, in short GDE framework. The GDE framework can be used as a basis for
evaluation of specific driving education methods and also as basis for developing new ideas. The
four hierarchical levels and their interplay was described above, but in Table 1 several examples of
specific contents of knowledge and skills at each level are presented. The second dimension of the
framework is formed by three goals for training: basic skills and knowledge, knowledge and skills
concerning risk increasing factors and skills for self-evaluation.
The first column is familiar from the traditional goals of driver training. Basic knowledge and
skills for vehicle manoeuvring and mastery of traffic situations are typical contents of driver
training. However, some of the contents on the highest two levels of the hierarchy are not typically
included in driver training curricula (GADGET, 1999). The contents of this column are related to
knowledge and skills needed in order to manage in normal traffic. On the lowest level this means
basic vehicle manoeuvring skills like control of speed and position, sufficient automation of gear
shifting etc. Knowledge about effects of vehicle properties (front wheel drive/rear wheel drive/four
wheel drive) and cargo are also included. Basic skills and knowledge required for mastery of traffic
situations are based on knowledge of rules, but also on skills for communication, speed adjustment,
observation etc. Basic skills on the second highest level, goals and context of driving are
connected with planning the trip, and with knowledge about the effects of time pressure and goals
of the trip on driving. In driver training the idea of the highest level, (and also of the second
highest level) would be to give the driver an idea of driving as a form of behaviour, in which
success or failure are closely related to a person’s motivational characteristics and the strategy he
or she chooses.
208 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
Table 1
GDE framework (GADGET, 1999)
Hierarchical level
of behaviour
Essential contents (examples)
Knowledge and skills Risk-increasing factors Self-evaluation
Goals for life and
skills for living
(general)
Knowledge about/control over how
life goals and personal tendencies
affect driving behaviour
Risky tendencies Self-evaluation/awareness
of
 lifestyle/life situation  acceptance of risks  personal skills for
impulse control
 group norms  self-enhancement
through driving
 risky tendencies
 motives  high level of
sensation seeking
 safety-negative motives
 self-control, other characteristics  complying to social
pressure
 personal risky habits
 personal values  use of alcohol/drugs
 values, attitudes
towards society
etc. etc. etc.
Goals and context
of driving
(trip related)
Knowledge and skills concerning Risks connected with Self-evaluation/awareness
of
 effects of trip goals on driving  driver’s condition
(mood, BAC etc.)
 personal planning skills
 planning and choosing routes  purpose of driving  typical goals of driving
 evaluation of requested driving
time
 driving environment
(rural/urban)
 typical risky driving
motives
 effects of social pressure in car  social context and
company
 evaluation of necessity of trip  extra motives
(competing etc.)
etc. etc. etc.
Mastery of traffic
situations
Knowledge and skills concerning Risks caused by Self-evaluation/awareness
of
 traffic rules  wrong expectations  strong and weak points
of basic traffic skills
 observation/selection of signals  risk-increasing
driving style (e.g.
aggressive)
 personal driving style
 anticipation of course of
situations
 unsuitable speed
adjustment
 personal safety margins
 speed adjustment  vulnerable road-users  strong and weak points
for hazard situations
 communication  not obeying rules/
unpredictable behaviour
 realistic self-evaluation
 driving path  Information overload
 driving order  difficult conditions
(darkness etc.)
(continued on next page)
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 209
The second column, risk-increasing factors, consists of knowledge and skills related to risks.
The contents of risk increasing factors are actually a part of knowledge and skills but they are
separated from other content because of their importance. The risks are different on different
levels of the hierarchy. The frequently-used concept ‘‘hazard perception’’ is a good example to be
analysed. By using the GDE framework it is easy to see that the traditional idea of hazard perception
as ‘‘road-craft’’ appears rather limited. There are potential hazards at all levels of the
hierarchy the driver needs to be able to recognize, such as risks related to type of the trip or
personal motives or behavioural tendencies.
This column contains topics, that are essential in the so-called ‘‘insight’’ learning courses (see
Gregersen, 1996b) as well as defensive driving courses. Defensive driving courses may emphasise
mainly skills based on vehicle manoeuvring when involved in a risky situation, and avoidance of
risks and risky situations is secondary (DAN, 2000). As for the first column, driver-training
programs probably do not contain as many topics connected with the highest two levels as with
the two lowest ones. It is evident, that insufficient knowledge of or skills related to all the aspects
of driving that are presented in the first column bear a possibility for increased risk and mastery of
skills and knowledge improves safety. However, the function of this column is to emphasise the
typical risk factors and describe these more in detail.
The content of the third column, self-evaluation, is an important tool in driver training and also
in development of driving skill after training. Research on development of expertise has shown,
that metacognitive skills and reflective thinking are essential characteristics of an expert (Kolb,
1984; Mezirov, 1981; Mezirov & Associates, 1990). Furthermore, reflective thinking is also an
essential part of development of an expert. Thus, training of self-evaluative skills should also be
included in the training, they do not develop automatically. Abilities for self-evaluation are also
expected to have relevance in driving behaviour. A driver, who is, for example, aware of his
tendency to fall asleep or his tendency to be unable to resist peer pressure, or a driver who knows
the limitations of his or her personal skills on slippery road conditions, may be able to take these
factors into consideration and adapt his or her driving accordingly. Educational methods, that
Table 1 (continued)
Hierarchical level
of behaviour
Essential contents (examples)
Knowledge and skills Risk-increasing factors Self-evaluation
 distance to others/
safety margins
 insufficient
automatism/skills
etc. etc. etc.
Vehicle manoeu-
vring
Knowledge and skills concerning Risks connected with Awareness of
 control of direction and position  insufficient automa-
tism/skills
 strong and weak points
of basic manoeuvring skills
 tyre grip and friction  unsuitable speed
adjustment
 strong and weak points
of skills for hazard situa-
tions
 vehicle properties  difficult conditions
(low friction etc.)
 realistic self-evaluation
 physical phenomena
etc. etc. etc.
210 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
might be appropriate to increasing driver’s skills for self-evaluation are improved feedback during
training, self-assessment tools like questionnaires and scales, discussions with other youngsters
about personal experiences and evaluations made by instructors or examiners.
Skills for self-assessment as well as driver attitudes and individual needs, which can be considered
as part of the highest levels in the hierarchy that are presented in the GDE framework,
have been referred to by Lynam and Twisk (1995) as the most promising aspects for finding
improvement in driver training. In this paper these are intertwined and seen in a wider perspective
and as a hierarchy. The GDE framework also gives a comprehensive tool for evaluating of driver
education methods along two dimensions at the same time.
It is hypothesised that driver education (at least pre-licensing) is still too concentrated around
the boxes in low-left corner of the GDE framework, while those in upper right corner tend to be
ignored and miss effective methods. Perhaps these methods are found in some areas of driver
education, e.g. driver improvement programs, or are developed for experimental purposes, but
they are not widely implemented in driver training. There is only a limited number of practical
educational methods, that are aimed specifically at the two highest levels in the hierarchy
(GADGET, 1999). Driving methods improving drivers’ metacognitive skills are equally scarce.
4. Conclusions
The evaluation of existing driver education methods, as well as defining new goals and new
methods should not only be based on empirical results but also on a conceptual analysis. Empirical
evidence is essential when developed or already implemented methods are tested, whereas
conceptual analysis helps in making new questions and finding the ‘‘white areas in the map’’. A
traditional criterion for a well-functioning driver training or driver improvement has been the
number of violations and the number of accidents after training. Driver training, especially formal
training has been criticised because of rather limited effects on accidents. This criticism is justified,
but not enough to prove the activity in itself useless. A fact is that there is no single technical,
legislational or behavioural means for reducing the number of accidents radically. Improvement
of safety is based on a wide net of countermeasures. It should be pointed out also, that if there is
no clear signs of good results with a certain driver training method it cannot be concluded, that
training in general is ineffective. In many studies concerning the effects of driver training, the focus
has been in the differences between formal and informal training, but the specific goals, contents
and methods of the formal training have not been analysed. The goals and the pedagogical
methods should be carefully analysed when an safety effects of training are evaluated.
The conclusions are presented as questions that arise from the GDE framework. The first
question concerns the framework itself. There are now four levels in the framework, but is that
sufficient? Should there also be a fifth level dealing with the phenomena and effects of society on
driver’s lifestyle, and with the possibilities for learning self-reflection or self-control? The driver is
not isolated but operating in certain circumstances and under certain conditions. Some researchers
have tried to integrate societal factors into their driver behaviour models (Michon, 1976)
or tried to get a grip on the latent, system errors in accident causation (Reason, 1990). The
question can be raised whether society is doing everything that it can and what it should do to
help the driver to drive safely?
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 211
The second group of questions concentrates on the goals and contents of training. Are all levels
of driver behaviour hierarchy emphasised enough in professional and layman instruction? Is
training still concentrating too much on the two lowest levels and too much limited to basic
knowledge and skills? Or should it concentrate on those two lowest levels and should training not
touch the highest levels but consider these as a person’s private areas. In particular trying to
change or affect the highest level can be seen as an intrusion to the persons private life and values.
However, in driver improvement courses the highest levels are in the main focus of work (DAN,
2000).
The third group of questions concerns the methods of driver education. What are the consequences
if learning of basic knowledge and skills, and concentration on vehicle manoeuvring and
mastery of traffic situations are not considered as a sufficient target of driver education? What are
the best methods to support learning when instruction aims at learning and understanding risks?
What are the pedagogical implications of trying to cover also the two highest levels in the GDE
framework? Is lecturing enough?
Self-evaluative or reflective skills were presented in the third column of the GDE framework.
What could be their role in the implementation of the idea of ‘‘life-long learning’’ in traffic? What
kind of measures are effective in teaching those skills or is it possible to teach them at all. Do we
have to support the learning process and how can we do it in an effective way?
It is also possible to discuss about the probable effects of different kind of training methods.
There is a possibility that a novice driver gets feedback concerning only about basic manoeuvring
skills and skills needed in mastery of traffic situations. What happens if the driver does not get
feedback or consider the role of the less concrete highest hierarchical levels before getting the
license? We already know what the problems are if training to drive on slippery roads is concentrating
on manoeuvring skills. We may also have other questions concerning driver training,
and systematic and conceptually well based tools help us to formulate the right questions and help
us to answer those in the future.
Acknowledgements
The content of this paper is mainly a result of EU-funded research project GADGET.
References
Bartl, G., & Stummvoll, G. (2000). Description of post licensing methods in Austria. In G. Bartl (Ed.), DAN-Report.
Results of EU-project: Description and Analysis of Post Licensing Measures for Novice Drivers. Vienna:
Kuratorium fur Verkehrssicherheit.
Bartl, G. (2000). DAN-Report. Results of EU-project: Description and Analysis of Post Licensing Measures for Novice
Drivers. Vienna: Kuratorium fur Verkehrssicherheit.
Bruhning, E., & Kuhnen, M. (1993). Strukturverkgleich des Unfallgeschehens in den neuen und alten Bundesl€aandern.
In Verkehrssicherheir im vereinten Deutschland. Mensch und Sicherheit. (Heft M 10). Bergisch Gladbach:
Bundesanstalt f€uur Straßenwesen.
Christensen, P., & Glad, A. (1996). Mandatory course of driving on slippery roads for driver of heavy vehicles. The
effect on accidents. (T€OOI Report 334/1996). Oslo: Transport€ookonomisk institutt.
212 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
Evans, L. (1991). Traffic safety and the driver. NY: Van Nostrand Reinhold.
Farrow, J. A. (1987). Young driver risk taking: a description of dangerous driving situations among 16- and 19-year-old
drivers. The International Journal of Addictions, 22, 1255–1267.
Fitts, P., & Posner, M. (1967). Human Performance. Belmont, CA: Brooks/Cole.
Fuller, R. (1984). A conceptualization of driving behaviour as threat avoidance. Ergonomics, 27, 1139–1155.
Fuller, R. (1988). Psychological aspects of learning to drive. In J. A. Rothengatter, & R. de Bruin (Eds.), Road user
behaviour: theory and research. Assen: Van Gorcum.
Glad, A. (1988). Fase 2 i f€ooreroppl€aaringen. Effekt paa ulykkes risikoen. (Rapport 0015). Oslo: Transport€ookonomisk
institutt.
Gregersen, N. P. (1996a). Young drivers’ overestimation of their own skill––an experiment on the relation between
training strategy and skill. Accident Analysis and Prevention, 28, 243–250.
Gregersen, N. P. (1996b). Young car drivers. (VTI Rrapport 409A) Link€ooping: VTI.
Gregersen, N. P., & Berg, H.-Y. (1994). Lifestyle and Accidents Among Young Drivers. Accident Analysis and
Prevention, 26, 297–303.
Gregersen, N. P., Berg, H.-Y., Engstr€oom, I., Noleen, S., Nyberg, A., & Rimm€oo, P.-A. (2000). Sixteen years age limit for
learner drivers in Sweden––an evaluation of safety effects. Accident Analysis and Prevention, 32, 25–36.
Hatakka, M., & Keskinen, E. (1999). Analysis of matrix. In G. Bartl, R. Cummins, C. Evers, M. Hatakka, E. Keskinen,
M. Pannacci, & G. Stummvoll (Eds.), Final report of European Commission project DAN. Vienna: Kuratorium fur
Verkehrssicherheit.
Hatakka, M., Keskinen, E., Katila, A., & Laapotti, S. (1994). Tell me about your car––I’ll tell you about your
driving habits! A paper presented in the 23rd international congress of applied psychology, 17–22 July 1994, Madrid,
Spain.
Janssen, W. H. (1979). Routeplanning en geleiding: Een Literatuurstudie. (Report IZF 1979 V.13). Soesterburg
(The Netherlands): Institute for Perception TNO.
Jessor, R. (1982). Problem behavior and developmental transition in adolescence. The Journal of School Health, 295–
300.
Jessor, R. (1987). Risky driving and adolescent problem behavior: an extension of problem behavior theory. Alcohol,
Drugs and Driving, 3, 1–11.
Job, R. (1990). The application of learning theory to driving confidence: the effect of age and the impact of random
breath testing. Accident Analysis and Prevention, 22, 97–107.
Jonah, B. (1986). Accident risk and risk-taking behaviour among young drivers. Accident Analysis and Prevention, 18,
255–271.
Jonah, B. (1990). Age differences in risky driving. Health Education Research, 5, 139–149.
Katila, A., Keskinen, E., & Hatakka, M. (1996). Conflicting goals of skid training. Accident Analysis and Prevention, 28,
785–789.
Keskinen, E. (1982). Human factor in traffic. University of Turku (in Finnish).
Keskinen, E. (1996). Why do young drivers have more accidents? Junge Fahrer Und Fahrerinnen. Referate der Esten
Interdiziplin€aaren Fachkonferenz 12–14 Dezember 1994 in K€ooln. Berichte der Bundesanstalt fur Strassenwesen.
Mensch und Sicherheit, Heft M 52 (in English).
Keskinen, E., Hatakka, M. Katila, A., & Laapotti, S. (1992). Was the renewal of driver education successful? Final
report of the evaluation project. University of Turku (in Finnish).
Keskinen, E., Laapotti, S., Hatakka, M., & Katila, A. (1994). How to lie with the concept of accident risk. A paper
presented in the 23rd International Congress of Applied Psychology Madrid 17–22.7.1994.
Kokkonen, M., & Pulkkinen, L. (1999). Emotion regulation strategies in relation to personality characteristics
indicating low and high self-control of emotions. Personality and Individual Differences, 27, 913–932.
Kolb, D. (1984). Experiential learning. Experience as the source of learning and development. Englewood Hills, NJ:
Prentice-Hall.
Laapotti, S. (1994). Fatal car accidents of novice drivers. Proceeding in the First Interdisciplinary Conference on Young
Drivers. 12–14 Dezember, Cologne, Germany.
Laapotti, S., & Keskinen, E. (1998). Differences in fatal loss-of-control accidents between young male and female
drivers. Accident Analysis and Prevention, 30, 435–442.
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 213
Laapotti, S., Keskinen, E., Hatakka, M., & Katila, A. (1996). Is the connection between quantity and quality of driving
and accidents similar for young and middle-aged drivers? A poster presented in the International Congress on Traffic
and Transport Psychology, 22–25 May 1996, Valencia, Spain.
Lewis, C. (1985). Adolescents’ traffic casualties: causes and interventions. In L. Evans, & R. Schwing (Eds.), Human
behavior and traffic safety. NY: Plenum press.
Lynam, D., & Twisk, D. (1995). Car driver training and licensing systems in Europe. (TRL Report 147). Crowthorne:
Transport Research Laboratory.
Marthiens, W., & Schulze, H. (1989). Analyse n€aachtlicher Freizeitunf€aalle junger Fahrer in Disco-Unf€aalle, Fakten und
L€oosungsstrategien. Bergish Gladbach: Bundesanstalt f€uur Strassenwesen.
Maycock, G., Lockwood, C., & Lester, F. (1991). The accident liability of car drivers (TRL-report 315). Crowthorne:
Transport Research Laboratory.
Mezirov, J. (1981). A critical theory of adult learning and experience. Adult Education, 32, 3–24.
Mezirov, J., & Associates (1990). Fostering Critical Reflection in adulthood. Jossey-Bass inc.
Michon, J. A. (1976). The mutual impacts of transportation and human behavior. In P. Stringer, & H. Wenzel (Eds.),
Transportation planning for a better environment (pp. 221–236). NY: Plenum Press.
Michon, J. A. (1985). A critical view of driver behavior models: what do we know, what should we do? In L. Evans, &
R. Schwing (Eds.), Human behavior and traffic safety. NY: Plenum Press.
Michon, J. A. (1989). Explanatory pitfalls and rule-based driver models. Accident Analysis and Prevention, 21, 341–
353.
Mikkonen, V., & Keskinen, E. (1980). The theory of internal models in traffic behaviour. General psychology
monographs, No B1. University of Helsinki (in Finnish).
Miller, G. A., Galanter, E., & Pribram, K. H. (1960). Plans and the structure of behavior. New York: Holt Rinehart and
Winston.
Mourant, R., & Rockwell, T. (1972). Strategies of visual search by novice and experienced drivers. Human Factors, 14,
325–335.
N€aa€aat€aanen, R., & Summala, H. (1974). A model for the role of motivational factors in drivers’ decision-making. Accident
Analysis and Prevention, 6, 243–261.
Parker, D., Reason, J., Manstead, A., & Stradling, S. (1995). Driving errors, driving violations and accident
involvement. Ergonomics, 38, 1036–1048.
Pulkkinen, L. (1998). A two-dimensional model as a framework for interindividual differences in social behaviour. In D.
Saklofske, & S. Eysenc (Eds.), Individual differences in children and adolescents (pp. 27–37). New Brunswik:
Transaction publishers.
Quimby, A., & Watts, G. (1981). Human factors and driving performance (LR 1004). Crowthorne: Transport Research
Laboratory.
Ranney, T. (1994). Models of driving behavior: a review of their evolution. Accident Analysis and Prevention, 26, 733–
750.
Rasmussen, J. (1980). What can be learned from human error reports?. In K. Duncan, M. Grunenberg, & D. Wallis
(Eds.), Changes in working life. London: Wiley.
Reason, J. (1990). Human error. NY: Cambridge University Press.
Reason, J., Manstead, A., Stradling, S., Baxter, J., & Campbell, K. (1990). Errors and violations on the roads: a real
distinction? Ergonomics, 33, 1315–1332.
Renge, K. (1983). Junge Kraftfahrer in Japan. (Forschungsberichte der Bundesanstalt f^uur Straßenwesen 92, Bereich
Unfallforschung). Bergisch Gladbach: Bundesanstalt f€uur Straßenwesen.
Rothengatter, J. A. (1997). Editorial preface. Applied Psychology, 46, 221–222.
Schulze, H. (1990). Lifestyle, leisure style and traffic behaviour of young drivers. In VTI-report 364 A. Swedish Road
and Traffic Research Institute, Link€ooping.
Siegrist, S. (Ed.) (1999). GADGET: An EU-project Guarding Automobile Drivers through Guidance, Education and
Technology. (WP 3 report). Bern: BFU.
Summala, H. (1985). Modeling driver behavior: a pessimistic prediction. In L. Evans, & R. Schwing (Eds.), Human
behavior and traffic safety. NY: Plenum Press.
Swedish Road Administration (1988). Swedish National Road Safety Report. Borl€aange, Sweden, 1998.
214 M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215
Twisk, D. (1994a). Improving safety on young drivers: in search of possible solutions. SWOV Institute for Road Safety
Research, manuscript.
Twisk, D. (1994b). Young driver accidents in Europe. Leidschendam: SWOV Institute for Road Safety Research.
Utzelmann, H., & Jacobshagen, W. (1997). Validation of the German system of diagnosis and rehabilitation for traffic
offenders. In J. A. Rothengatter, & E. CarbonellVaya (Eds.), Traffic and transport psychology, theory and application.
Oxford: Pergamon.
Van der Molen, H., & B€ootticher, A. (1988). A hierarchical model for traffic participants. Ergonomics, 31, 537–555.
Vogel, R., & Rothengatter, J. A. (1984). Motieven van snelheidsgedrag op autosnelwegen; een attitude onderzoek.
Traffic Research Center Report VK 84-09. Haren: Traffic Research Centre (in Dutch).
Wikman, A.-S., Nieminen, T., & Summala, H. (1998). Driving experience and time-sharing during in-car tasks on roads
of different width. Ergonomics, 41, 358–372.
Williams, A. F., & O’Neill, B. (1974). On-the-road driving records of licensed race drivers. Accident Analysis and
Prevention, 6, 263–270.
M. Hatakka et al. / Transportation Research Part F 5 (2002) 201–215 215